WormBase-Caltech Weekly Calls October 2016

October 6, 2016

Citace upload

• Tuesday, October 25th, 2016

Automated descriptions

• Ranjana working with Juancarlos to run and update automated descriptions pipeline

Intellectual lineage

• First draft up on WormBase development site

Noctua

• Raymond trying to annotate axioms for anatomy function
• How to connect cell types in different organisms to those in C. elegans anatomy ontology?
• Maybe best way is to connect based on biological function, e.g. regulate osmolarity (kidney, excretory cell)
• Raymond attempting to use Noctua to draw relationships between anatomy terms and function
• There's a distinction between annotating to an instance vs. to a class
• Raymond worked on modeling a statement on excretory cell function from WormAtlas
• Question: how much redundancy might we end up with if we annotate to higher level (cell/anatomy function) and lower level (individual gene product function)? There may be, but all of this information is valuable

Expr_pattern model change

• Daniela proposed a CRISPR-Cas9 tag for the expression pattern class as an experiment type
• Would be inline with "Reporter gene", "In situ", "Antibody", etc.
• Probably want "Genome editing" instead of "CRISPR-Cas9" so as to be more generic
• Would a GFP knock-in via genome editing be distinct from "Reporter gene"?
• Could have a super tag "Genome_editing" and sub-tags "CRISPR-Cas9", "Zinc_fingers", "TALEN"
• Maybe we'll just stick with a "Genome_editing" tag

October 13, 2016

Strain requests

• Ranjana sending strain requests to Mary Ann
• Mary Ann processing published strains requested by Ranjana

GO meeting

• Registration is now setup
• Sarah (Torres) could register all WB CIT participants
• Kimberly is helping organize

AGR Portal Use Case working group update

• Working on first use case, basic text search for genes, GO terms, diseases, and orthology groups
• Data is being collected into a "Portal Data" Google drive: https://drive.google.com/drive/u/1/folders/0B9qQtJIEUDEhRnJKWF9NTTVMVGs
• Some MODs providing more data than requested; That's OK, we just want to keep it simple for the first implementation
• We only want, for the first pass, to retrieve genes, GO terms (no GO-gene annotations), diseases (no disease-gene annotations), and orthology groups
• We likely already have enough data for a first pass implementation; not comprehensive data, but that is fine for now
• Disease data dumped from ACEDB, based on Disease Ontology; not comprehensive for OMIM disease list
• We'll eventually want to pull data directly from OMIM and GO

Extending SPELL virtual machine

• Has a lot of data; may be contributing to SPELL downtime
• Could possibly move to the cloud, but we would need a machine in the cloud
• Can only install SPELL on Ubuntu 8; SGD uses RedHat; we can ask them what they're using, how it's working
• Possibly SGD could run a worm instance on their machine(s)

Transgenome

• Junacarlos finished parsing the transgenome data with Daniela
• Checking parse output numbers with the Transgenome group
• Will the expression data be dumped out in the expression GAF? Yes, should be
• Data from Transgenome are not necessarily peer-reviewed; need to make that clear in display
• Some Expr_pattern objects can be linked to WBPerson, if not a WBPaper
• Data in GAF will point to the 2012 Cell paper as a reference for the Transgenome project
• Raymond suspects issues with it; what if users cannot trace data back to source (by looking up the paper)?
• Would be best to link directly to the respective Transgenome database page when displaying their data
• Notes can be added to release notes that the Transgenome data has been added to the database

Micropublications

• Trial with uploading Nemametrix data; aiming in putting data for the next upload (2-4 micro)
• Added micropublication type in the ?Paper class
• Micropublications will acquire WBPaper IDs
• Previous micropublications were only ?Expr_pattern objects; now will be ?Paper objects
• Ideally micropubs. will be indexed by PubMed and enter the WB pipeline through the normal paper route
• Question: create WBPaper objects as soon as accepted? That is plan for now
• Micropubs. will first be hosted by WormBase, but may move to an independent repository (Micropublication: biology)
• Will we display images/figures from micropubs. only or for all papers? We'll need to wait and see
• Could we accommodate any digital data to associate with the micropub.? E.g. data points, plasmid maps, etc.
• Micropubs. text will be marked up as much as possible; what about making all sentences OWL statements?
• Would this be changing the way that authors write papers? Maybe, but should be independent of core micropublication pipeline
• Each micropublication/data/result could be combined as modules to create a larger, more comprehensive narrative
• There is a notion of a "millipublication" that could be a larger narrative from collected micropublications
• One original idea was that micropublications would capture data that people would not otherwise want to publish
• Data submitted as micropublication will not be able to be submitted as another publication later but will be cited as data point
• Micropubs. will be considered primary research article, not just an abstract (equivalent to meeting abstract)

October 20, 2016

AGR search prototype

• URL: http://search.alliancegenome.org/
• Searching "WNT4", then choosing category "Gene Ontology" provides facets for species
• Question is: how are these species associated to the GO terms?
• Associated genes could have a species prefix or indicator

AGR gene page

• There was resistance to creating AGR gene pages
• Some see it as redundant with what is already available elsewhere
• Although we may want to build a foundation gene page and develop from there
• Maybe MODs could develop their respective gene pages to look more like an AGR standard
• One main reason to create AGR is to have a consolidated, consistent, standard view for data, including genes
• Would be good to assemble a user group (or intended audience) to receive feedback from
• Use Case working group has put together some ideas for a gene comparison page, keyed off of orthology group:
  • https://docs.google.com/presentation/d/1frMMh3xfepesJq5iqSXAwwLOrb45cuqe3LValEOd4bM/edit#slide=id.p

Citace upload

• Next Tuesday, October 25th 10am

Curator call next week

• Topic: curation styles
• Curation by paper, by topic, etc.
• Curation inter-dependencies
• Daniela will present for WormBase
• Send ideas to Daniela for her to cover
• Send older slides with relevant information to Daniela

October 27, 2016

AGR Portal Use Case working group

• Working group had call yesterday with Paul Thomas and Judy Blake
• Decision was made to abandon the idea of displaying "orthology groups"/"homology groups"
• Instead we will display a gene-centric page with comparisons of focus gene data with data from all orthologs and paralogs
• Paul Thomas' group had built a prototype for pulling out orthologs with GO info: http://umodtest.usc.edu/
• This prototype returns closest orthologs of a focus/query gene and then displays a table of all GO annotations for all genes
• We would want to display orthologs for a focus gene as well as paralogs
• This display is very similar to the gene comparison page mockup the use case working group had been envisioning, except now the focus will be a single gene, not a PANTHER subfamily

Disease-Strain curation

• Ranjana working on disease-strain associations for WormBase
• Disease group working on the DAF (disease association file) format
• Ranjana will discuss the strain curation pipeline with Mary Ann and Karen
• Will want a unique ID for each strain, especially for when we merge strain data with other MODs
  • One solution is to just prefix the strain name with "WBStrain:" to indicate it came from WormBase

Kimberly visiting CA

• Kimberly will be at CIT next Wednesday
• Noctua developers will be around to discuss integration with Textpresso
• Kimberly can send around schedule for the GO meeting
• Would be good for curators to setup a Noctua account before the Noctua workshop (on Nov 7)
• Kimberly will also be at CIT on Tuesday Nov 8th

Paper editor CGI

• Michael had some issues with the paper editor
• Seems to be working now